Observer variability in upper gastrointestinal endoscopy

Cantor, David; Imperato, T J; Midgley, Robert; Yassinger, S

doi:10.1016/0016-5085(78)93458-3

Cited by 2 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These patients were managed by a large pool of endoscopists, and thus outcomes are not reflecting the practice of a few. The lack of variation between the outcomes for individual endoscopists or for hospitals suggests that potential variation in the reporting practices of endoscopists did not have a significant impact upon the results [29][30][31] and lends support to the robustness of the study's findings. Limitations of the study included the absence of results of further endoscopic examinations which would have provided additional information regarding pathological changes.…”

Section: Discussionsupporting

confidence: 51%

A 9 year prospective cohort study of endoscoped patients with upper gastrointestinal symptoms

Westbrook

Duggan

Jm³

et al. 2005

Eur J Epidemiol

View full text Add to dashboard Cite

Despite the high prevalence of upper gastrointestinal symptoms and associated costs of diagnosis and management, evidence regarding long-term outcomes is scant. We studied symptom outcomes 18 months (FU1) and 8/9 years (FU2) post-index endoscopy to identify demographic, diagnostic and treatment factors associated with outcomes. A retrospective review of medical records at two Australian teaching hospitals identified a cohort of 302 patients who had an index endoscopy (performed by 23 endoscopists) 18 months previously. Patients were interviewed at FU1 and FU2. In total 34% (95%CI: 29.0 39.8) of patients were asymptomatic at FUI and 41% (95%CI: 35.6-46.6) at FU2. For 63%, outcomes at FUI predicted long-term outcome, with 19% (95%CI: 14.6-23.4) asymptomatic and 44% (95%CI: 38.4-50.0) symptomatic at both times. Those whose symptom status changed were as likely to deteriorate as improve (p > 0.05). Number and severity of presenting symptoms (F = 3.3, df = 3,277, p < 0.05) and older age (F = 2.8, df = 2,301, p < 0.05) were associated with poorer outcomes. Long-term outcome was unrelated to endoscopic diagnosis. Those symptomatic were significantly more likely to be on proton pump inhibitors (PPIs) or Histamine2 Receptor Antagonists (H2RAs) than those who were asymptomatic. Use of PPIs at FU2 was associated with a significantly better outcome than use of H2RAs. However this impact was relatively small, with 69% of patients on PPIs and 84% on H2RAs symptomatic at FU2. Upper gastrointestinal symptoms prompting endoscopy are chronic for the majority of patients regardless of diagnosis. Endoscopic diagnosis is of limited value in predicting long-term outcomes. The association between poor outcome and use of H2RAs and PPIs challenges views about their long-term effectiveness in symptom control.

show abstract

Section: Discussionsupporting

confidence: 51%

A 9 year prospective cohort study of endoscoped patients with upper gastrointestinal symptoms

Westbrook

Duggan

Jm³

et al. 2005

Eur J Epidemiol

View full text Add to dashboard Cite

show abstract

“…[15][16][17][18][19][20] For example, Barrett's esophagus was a common source of disagreement in this study, and the difficulty in reaching a consensus on diagnosis is highlighted in the literature where the diagnosis of low-grade dysplasia in Barrett's esophagus yielded kappa values indicating at best fair agreement beyond chance. 15 Moreover, in 1 study evaluating the reliability of UGI endoscopy between experts in which only major changes were compared, the reported interobserver agreement (across a number of disease categories) was only 32%, 16 with a separate study documenting 1.19 AE 1.08 major discrepancies and 2.18 AE 1.32 minor discrepancies.…”

Section: Discussionmentioning

confidence: 84%

“…15 Moreover, in 1 study evaluating the reliability of UGI endoscopy between experts in which only major changes were compared, the reported interobserver agreement (across a number of disease categories) was only 32%, 16 with a separate study documenting 1.19 AE 1.08 major discrepancies and 2.18 AE 1.32 minor discrepancies. 17 The lack of consensus on diagnostic terminology, both in this study and the literature, highlights the need for improved synoptic reporting language. Although the MST Diagnosis List for the Upper Gastrointestinal Tract (http://www.omed.org/index.php/resources/re_mst/) was extremely valuable, it introduced some undesirable variation in the study.…”

Section: Discussionmentioning

confidence: 92%